Date of Award

Winter 1-14-2026

Document Type

Thesis (Master's)

Department or Program

Engineering Sciences

First Advisor

Douglas Van Citters

Second Advisor

John-Erik Bell

Third Advisor

Solomon Diamond

Abstract

In vivo wear of ultra-high molecular weight polyethylene (UHMWPE) orthopedic bearings represents a major problem in shoulder and knee arthroplasty, with significant amounts of devices failing due to the complication itself, and even more due to associated reasons for retrieval, including aseptic loosening. However, quantification of UHMWPE wear has largely been limited to linear and in vitro volumetric analyses, obscuring the understanding of the amount of clinically relevant material loss. This study proposes new methods to measure articular, rim, and backside surfaces of retrieved shoulder and knee UHMWPE liners, centered around the coordinate measuring machine (CMM)-based collection of high density point clouds. Subsequently, the clouds are algorithmically manipulated using thresholding, least squares fitting, interpolation, and smoothing to regenerate unworn surfaces, which are compared to the worn surfaces to identify and calculate volumetric wear. Utilizing gravimetric analysis as a standard, both methods operated within an error of 9.0% (6.5 mm^3).

As a proof of concept for the methods’ potential clinical utility, they’re applied to characterize the relationship between in vivo wear and surgical, design, and material-based factors. Fifty reverse total shoulder arthroplasty and twenty total knee arthroplasty with varying retrieval reasons, in vivo durations, and manufacturers were analyzed. From the former, as already studied in hip arthroplasty, it was found that irradiation crosslinking in conjunction with thermal stabilization or antioxidant additive decreases in vivo articular and rim wear rates with statistical significance; however, no relevant differences were found between manufacturer designs. The latter preliminarily suggested possible differences in frontside versus backside wear for certain liner designs, but a larger sample size is necessary to make statistically-backed claims. Ultimately, as they relate to failure analysis and postmarket surveillance, these results suggest the potential clinical utility of generalized strategies that take point clouds as inputs— regardless of their collection method and implant type— and quantify in vivo material loss. Future efforts seek to understand more nuanced wear relationships, such as the effect of mechanical alignment on wear for knees, or the effects of center of rotation lateralization and neck shaft angle on wear for shoulders.

Download

Share

COinS